Open Collections

L £ 3 fi>? ? • / STAPHYLOCCUS EUTEROTOXIU by Agnes E. Mehling A thesis submitted in partial fulfilment of the requirements for the degree of Master of Arts in the department of Bacteriology and Preventive Medicine The University of British Columbia April, 1950 STAPHYLOCOCCUS ENTEROTOXBT The constituents of a synthetic medium and the conditions required "by staphylococcus strain 12069 alpha for good toxin production have been determined. Because essentially a l l the components of the medium are dialyzable, dialysis for 24 hours was found to produce an average re-duction of 98.34 per cent of the solids, but an average reduction of 53 per cent of the potency. Various methods of concentrating the toxin were studied. Vacuum d i s t i l l a t i o n experiments indicated an a l -most complete loss of potency. "Desi-vac" treatment pro-duced a loss of approximately 50 per cent, and a product which varied in weight content, st a b i l i t y , toxicity, and response to purification. Iyophilization produced a loss of approximately 20 per cent, and a product which reacted more consistently to further treatment. The freezing meth-od of concentration was shown to be applicable to the problem. The concentrated toxic f i l t r a t e s were subjected to additional purification procedures. The toxin was ad-sorbed from solution by "Horit", but i t s elution was not achieved. Sthanol, methanol, and acetone precipitation experiments were performed, but only the methanol treatment showed any evidence that would warrant futther investigation. Cadmium chloride precipitation removed the active compon-ent from f i l t r a t e s , hut residual cadmium proved too toxic in animal tests to permit the degree of separation of entero-toxin to be determined. The probable polysaccharide nature of enterotoxin was demonstrated by enzyme experiments upon crude toxic f i l t r a t e s , but further study was prevented because of the extremely small yields of relatively purified material. Enterotoxin purified by acid precipitation at pH'3.5 gave a negative Biuret and a positive Molisch reaction. The low antigenic power demonstrated by the relatively pure f i l t r a t e s was also indicative of the non-protein nature of the toxin The acid precipitation at pH 3.5 of dialyzed and concentrated f i l t r a t e s produced a material whose soluble fraction was found to have an average weight of 150 gamma per cat test dose, and a nitrogen content of less than 2 per cent. The recovery of enterotoxic potency in material treated by dialysis, concentration, and precipitation, averaged 7.33 per cent of the original. ACKNOWLEDGEMENT I would like to express my appreciation of the encouragement and advice which Dr. Dolman has offerred so generously throughout the course of this research. I would also like to thank the many members of the Department for their help and coop-eration, without which this work would have been impossible. TABLE OF CONTENTS STAPHYLOCOCCAL ENTEROTOXIN I. Introduction page 1 II. Production of Toxin A. Preparation of Synthetic Medium 5 B. Preparation of Filtrates 12 III. Purification of Toxin A. Dialysis 19 B. Concentration Methods 27 C. Precipitation 56 D. Adsorption 59 E. Freezing 59 IV. Nature of Toxin A. Potency 60 B. Non-Protein Properties 61 LIST OF CHARTS Chart I. A summary of the various terms used in the acid precipitation technique of purifying DDC toxin. 29 Chart II. The percent of the original potency of C toxin to he found i n the various fractions involved in acid precip-itations 37 Chart III. Summary of experiment 2 : the effect of vacuum d i s t i l l a t i o n on DC toxin 48 LIST OF TABLES TABLE I: The completeness of removal of type C medium hy 42.5 hrs. dialysis through Visking «,No-JaxM tubing. page 22 T2LBE1 II: The removal of diffusible materials from type C toxin by dialysis through Visking "No-Jax" tubing page 22 Ti&BLE III: Removal of diffusible materials from type C toxin by dialysis through Visking »Hb-JaxM tubing page 24 TABLE IV: Comparison of C and DC toxins with respect to enterotoxic potency. page 25 TABLE V: The relation of the solid content of RS material to cat Test Doses page 31 TABLE VI: The relation of the solid content of SP material to cat test doses page 32 TABLE VII; The gradual removal of impurities as revealed by total solids content per cat test dose. ....page 33 TABLE VIII The per cent of original DDC solids found in the RP fraction page 33 TABLE IX: Determination of average percentage of DC cat test doses recovered in the various fractions page 35 TABLE X: The percentage of the original C toxin C.T.D.'s detected in the various fractions page 36 TABLE XI: The percent of the original DC potency recovered in individual fractions page 38 TABLE XII: A comparison of four bottlesof #26 DDC toxin which had undergone "Desi-vac " treatment. .page 42 TABLE XIII Determination of the average percentage of DC cat test doses in the various fractions of material treated by lyophilization page 5A TABLE XIV: The percentage of the original cat test doses in C toxin detected in the various fractions of lyophiliz-ed material page 52 TABLE XV: The percentage of the original cat test doses in the various fractions of material concentrated by lyoph-i l i z a t i o n at the C stage page 53 TABLE XVI: A comparison of the results obtained from 3 methods of concentrating DC toxin page 55 TABLE XVII. The recovery and degree of purity of ethanol precipitated toxin compared to SP. • page 57 TABLE XVIII. Milligrams of nitrogen in the var-ious f i l t r a t e s as revealed by micro-Kjeldahls page 62 1 STAPHYLOCOCCUS ENTEROTOXIN Introduction Staphylococcus enterotoxin i s undoubtedly res-ponsible for a greater number of food poisoning outbreaks than any other agent. But, because of the transient nature of the attack, the rapidity of complete recovery, and the frequency with which such attacks are experienced - this source of great discomfort and economic loss i s often dis-regarded, and only the more dramatic types of food poison-ing outbreaks are noted. Fortunately, i n recent years, there has been a growing awareness of the fallacy of this attitude, and a number of workers have turned their atten-tion to this product of the ubiquitous staphylococci. The study of this toxin was retarded for years by the belief that the presence of contaminating staphy-lococci in foodstuffs was usually irrelevant. The prevalence of these organisms* and the fact that only a few of the strains are capable of producing enterotoxin, may have done much to substantiate this belief in the minds of early investigators. Indeed, i t was not u n t i l 1930, when Dack et a l (1) f i r s t reported the administration of f i l t r a t e s to human volunteers* that the potential importance of these organisms was realized. Moreover, the contentious reports concerning the toxigenic spectrum of staphylococci added greatly to the confusion. The fact that both alpha and beta toxins apparently evoke some gcuteoga-stro.iintest-- 2 -inal symptoms found in enterotoxin poisoning led many workers to believe that they were dealing with onejsubstance. As recently as 1943, Pulton (2) reported the inseparabil-ity of enterotoxin and alpha-toxin, and in 1939, Kojima and Kodama (3)^and i n 1941, Woodward and Slanetz (4) id-entified enterotoxin with beta-toxin. However, certain physical and biological proper-ties have now established the fact that enterotoxin i s a distinct entity, and that i t i s only coincidental that many food poisoning strains produce alpha-orbeta- toxins as well. The comparative thermostability and low antigen-ic properties are characteristics of enterotoxin. The fact thst i t s effects are not n u l l i f i e d by antisera which neut-ralize alpha-and beta-toxins seems conclusive. Furthermore, a staphylococcus strain has been found which produces alpha-and enterotoxin to a high t i t r e under onejset of conditioss, but only negligible traces of alpha-toxin, and enterotoxin to i t s former level, under altered conditions. These pec-uliar properties, and others which w i l l be revealed in the course of this report, are indicative of the unique pos-it i o n that staphylococcus enterotoxin should be assigned. In addition to the contentious nature of early findings in this field , the problem of testing for enter-otoxin has always presented a handicap to the investigation of this subject. In spite of numerous attempts, no satis-factory method for determining tonicity has been found 3 which does not involve the pse of some laboratory animal -usually the cat, although in some cases, resort has been made to monkeys, or even humans* Thus, because of the particularly unpleasant effects of enterotoxin, the indef-initeness of the reaction which involves long hours of care-f u l observation, the aesthetic barriers presented by the usual cat test, and the d i f f i c u l t i e s experienced in the maintenance of a healthy supply of these animals - i t i s not surprising that investigations of this subject have been relatively rare. It i s hoped that with continued study of the problem, simplified tests w i l l be evolved. Once i t i s realized that enterotoxin exerts such an important and unavoidable influence on our daily lives, the necessity becomes apparent for data concerning the requirements of toxin production. If these be known, improved handling of food, with the elimination of a l l possible opportunities for enterotoxin formation, may re-sult in a happier enjoyment of a greater variety of food-stuffs. Obviously, in an investigation of this problem, one of the requirements for further study i s the develop-ment of a large-scale method of producing and purifying the toxin. As w i l l be apparent from this report, the extreme dilution of the potent factor as compared with oth-er toxins which have been purified, poses one of the most d i f f i c u l t problems of purification; and u n t i l a dependable method, adaptable to large-scale production has been dev-4 ised, continued research w i l l be d i f f i c u l t . 1 In summary, the purpose of t h i s i n v e s t i g a t i o n has been: 1) to discover a d d i t i o n a l f a c t s concerning the nature of enterotoxin, so that further proof as to i t s p o s i t i o n as a d i s t i n c t entity w i l l be available, and so that a possible explanation may be found for i t s low antigenic powers and i t s s p e c i f i c attack on one centre of the central nervous system. 2) to study the conditions of toxin production so that a better understanding of essential food control problems may be obtained.. 3) to devise a method by which enterotoxin can be pro-duced and p u r i f i e d oh a large scale so that i t s further study w i l l be enhanced. II . PRODUCTION Off TOXIN Throughout t h i s i n v e s t i g a t i o n staphylococcus s t r a i n 12069 alpha was employed. This organism has proven to be consistently alpha- and enterotoxigenic. When grown on proteose peptone, semisolid agar medium for 40 hours at 37°C. i n an atmosphere containing 30 per cent carbon dioxide, this s t r a i n produces both toxins; however, when grown on the same medium at room temperature, only neg-l i g i b l e amounts of alpha toxin, but equal t i t r e s of enter-otoxin, are produced. (5) Thus, th i s p a r t i c u l a r staphylo-5 coccus strain affords a most suitable source of enterotoxin. By this simple switch of incubation temperatures, the f i r s t step of purification can be achieved; namely, the report-edly impossible separation of enterotoxin from alpha toxin. Whenever, during the course of experimental work, non-toxic f i l t r a t e s were obtained - the organism was always re-checked for possible variation by growing i t again on this soft agar medium, and at no time did negative f i l t r a t e s result under these conditions. A. Preparation of Synthetic Medium As soon as the study of enterotoxin was undertaken, the necessity for a synthetic medium became apparent. A chemically reproduceable medium was essential in order that consistent toxin production could be assur/ed. But of greater importance was the fact that a less comples med-ium would simplify greatly the purification procedure. The advantage of having the constituents of the medium as unrelated as possible in molecular size and behaviour to the material to be purified was recognized. From ear-ly work concerning the possible nature of enterotoxin, i t was evident that i f this substance were a protein, i t was probably of the molecular size of the proteoses ( 6 ) ; but that i t might even be a complex carbohydrate (7). Thus the inadvisability of the continued use of proteose peptone semi-solid agar was obvious. The nutrition of staphylococci in general, has 6 been studied in considerable detail, and a perusal of the literature produced an excellent background of information. The work of Fildes, Richardson, Knight, and Gladstone ( 8 ) revealed the essentiality of the following amino acids for the rapid growth of staphylococci: arginine, aspartic acid, cystine, glycine, histidine, leucine, phenylalanine, proline, and valine. Lysine, methionine, and tyrosine were of variable importance, and tryptophane was required by some strains. The only essential vitamins according to Knight (9) were thiamin and nicotinic acid, but Porter and Pelczar (10) reported biotin to be required by certain strains. The necessity of uracil for anaerobic growth was noted (11). Fildes, et a l (8) found that ferrous iron and magnesium satisfied the mineral requirements, that d-glucose furnished a convenient carbohydrate source, and that phos-phate served as an excellent buffering agent. After preliminary work u t i l i z i n g the simplified medium of Favorite and Hammon (12) and a medium designated "type M", Casselman (13) devised a synthetic medium (type C) for use in the study of enterotoxin. A l l the prelimin-ary experiments involved i n the evolution of this medium concerned themselves only with maximum growth, and disre-garded the subject of maximum toxin production. Because later work has shown that toxicity does not necessarily parallel growth, i t i s quite possible that by attention to this latter fact, a medium could be devised which would result i n greater toxin production. However, d i f f i c u l t i e s involved in the testing for toxin injthe innumerable poss-i b i l i t i e s * and the fact that the toxin level on the "C" medium was originally found to be almost as high as on the soft agar - led to the acceptance of type "C" medium as the basis for enterotoxin production by strain 12069 alpha. But**, the frequent appearance of slimy or "rough « growth, the lowering of potency levels to less than half of earlier ones from time to time, and the occasional per-iod when f i l t r a t e s revealed no toxicity at a l l , suggested that type "C" medium does not enhance the s t a b i l i t y of enterotoxigenic properties of this organism. At the beginning of this research, this type C medium was prepared exactly as outlined by Casselman, but later, because of reasons to be detailed shortly, minor modifications were made. A summary of the preparation of C medium follows. A. Chemicals and Solutions Required. 1. Solution of Bacto-Casamino Acids prepared accord-ing to procedure on next page. * - 2 . Bacto-dextrose i n d i s t i l l e d water, 50$ 3 . Potassium dihydrogen phosphate (CP.) in d i s t i l l e d water, 10$ 4. Sodium hydroxide (C.P.) in d i s t i l l e d water, 10$ * 5. Nicotinic acid in d i s t i l l e d water, 0.5% *" 6 . Thiamin hydrochloride in d i s t i l l e d water, 3 . 3 $ * 7. Uracil i n d i s t i l l e d water, 1.0$ ** 8. Mineral mixture prepared, "by dissolving the following C.P. salts in 1 l i t r e of d i s t i l l e d wat-er and adjusting the mixture to pH 3.0 or lower with concentrated HCl MgSOy.. '7EKQ 20 gms PeSOf. 711^ 0 6.0 gms MnSO^ .^ H^ O 2.0 gms 9. Disodium phosphate (C.P.), solid * These solutions were sterilized hy autoclaving in large flasks at 240 P. for 10 mins. B. Preparationbf Solution of Casamino Acids. A solution of 500 gms of Bacto-Casamino Acids in 5 l i t r e s of distilledjwater was prepared and adjusted to pH 3.5 with concentrated hydrochloric acid. Approximately 100 gms. of Eorit-A was added to the acidified solution, and the mixture was stirred well to disperse the decoloriz-ing charcoal. After being allowed to stand for 20 minutes i t was f i l t e r e d through Whatman's No. 1 f i l t e r paper on a Buchner funnel. If the filtrate;, was at a l l coloured, the treatment with Eorit-A was repeated after rechecking the pH. To the colourless f i l t r a t e were added: arginine hydrochloride 15.0 gms. cystine 3.5 gms. glycine 20.0 gms. tryptophane 2.5 gms. The mixture was stored i n suitable flasks i n the refriger-ator. 1000 ml. 3750 ml. 2.5 ml. 2.5 ml. 3750 ml. 200 gms. C. Preparationjof Medium, part (1) solution of Bacto-Casamino Acids d i s t i l l e d water nicotinic acid (0.5%) u r a c i l (1.0%) NaOH (10%) - to pH 7.4 part (2) d i s t i l l e d water Na^HPO^ .12HjO warmed to dissolve . part (3) mineral mixture thiamin HCl (3.3$) prepared with aseptic precautions 1) Parts (1) and (3) were mixed and adjusted to pH 7.4 with 10$ KH^ PO^ . 2) Sufficient water was added to make the volume up to 9950 ml. 3) If necessary, the mixture was readjusted to pH 7.4 with 10$ KH^ PO^ . or 10$ NaOH and then the volume was made up to 10.0 l i t r e s . 4) This mixture was dispensed originally 1 l i t r e per gal-lon bottle and autoclaved at 252 P. for 20 mins. 5) The following were added to each bottle when cool aid 34 ml. 2 ml. 10 -before inoculation: part (3) 3 ml. 50$ d-glucose 20 ml. It is interesting to note the similarity between this type C medium and that devised by Surgalla (14 & 15) for enterotoxin production. The ingredients are much the same, the major difference being his use of individual am-ino acids throughout; whereas, type C medium u t i l i z e s Bacto-Casamino acids as i t s base. Moreover, in general, the concentrations of the substanc es are higher in C medium, and a few notable examples follow: thiamin i s added in ap-proximately 200 times the quantity Surgalla uses; glucose, 5 times; cystine, 5times; and so on. The inclusion of manganese in type C medium i s another differentiating point. Unfortunate^ , consistently potent batches of enterotoxin were not obtained, and Wood (16) reported neg-ative cat tests in quantities up to 2.5 cc. for some 6 weeks before the problem was taken over by this worker. When 2 additional batches o f . f i l t r a t e produced negative cat reactions in quantities up to 3.0 c c , attention was turned again to the constituents of the medium and the conditions of growth. It was realized that variation of the organism i t s e l f might be involved, but when f i l t r a t e s prepared on proteose-peptone, semi-solid agar yielded the usual potenc-ies of enterotoxin - this possibility was discredited. 11 The variation was one which developed only when the organ-ism was grown on C medium. At a l l times* growth in the non-toxic cultures was just as abundant as i t had been in toxic ones. Because the switch" from 4 per cent to 2 per cent disodium phosphate had been made by Casselman (13) just prior to the time when the non-toxic f i l t r a t e s were obtain-ed, this factor was investigated f i r s t . But the increase of the phosphate content to the former 4 per cent value produced no detectable difference in the f i l t r a t e s obtain-ed. Next followed an investigation of the minerals us-ed. Was the inclusion of MnS04 in the medium ju s t i f i e d when one considered the former acceptance of the report of Fildes, Richardson, Knight, and Gladstone (8) that "the mineral requirements of the staphylococci are satisfied by ferrous iron and magnesium," and the subsequent finding of Surgalla (15) that enterotoxin i s produced on a synthetic medium containing only these 2 minerals? Could the manganese be exerting an inhibitory effect on enterotoxin production by 12G69 alpha under certain conditions? In his report (13), Casselman gave no explanation for i t s inclusion. However, when 12G69 alpha was grown on C medium, complete in every way except for the omission of manganese sulfate, non-toxic f i l t r a t e s were s t i l l produced. Later, Casselman, i n per-sonal communications, substantiated these findings by 12 stating that, after the appearance of Wolf's (17) paper, he employed manganese sulfate experimentally at various : times and found that concentrations as high as 0 .G004M (20 times the concentration in type C medium) had no inhibitory effects. The possible importance of biotin was also in-vestigated, because of the findings of Porter and Pelczar (10) that i t was essential for the growth of some strains of staphylococci, and because of i t s omission from G medium on the basis of growth, not toxin tests. However, the in-clusion of biotin produced no improvement i n the toxin pro-duction. During this study of the possible effects of phos-phate concentration, the omission of manganese, and the inclusion of biotin - not only the potency of the f i l t r a t e s , but also the degree of growth, colonial characteristics of the organisms when plated out, and the pH of the f i l t r a t e s , were investigated. However, at no time, was any detectable variation evident. At this point, attention to methods of production (which w i l l be summarized in the following sect-ion) resulted in the production of potent f i l t r a t e s , and the investigation of the effects of medium eonstituents on toxin production was discontinued. 33. Preparation of Filtrates The preparation of enterotoxic f i l t r a t e s always 13 commenced with the inoculation of a t y p i c a l colony of 12069 alpha from a 24-hour blood plate into 15 or 20 cc. of C medium. This was incubated at room temperature u n t i l growth became evident (usually 6 hours), and then t h i s entire culture was inoculated into an erlemneyer f l a s k containing s u f f i c i e n t C medium to provide a 2 per cent inoculum for the batch to be prepared. This f l a s k was well agitated at the time of inoculation, and l e f t at room temp-erature usually for 18-i&o24 hours - or at l e a s t u n t i l a f a i r l y turbid culture was obtained. Because of the f a c t that growth was enhanced i f the culture was shaken occas-i o n a l l y , and because of the l a t e r discovery of the influen-ce of aeration on enterotoxin production - the culture was c a r e f u l l y swished around from time to time. Por the same reason, an inoculating culture was never prepared i n a f l a s k which contained G medium to a depth of more than l-§- inches. At the beginning of t h i s project, the batches of toxin were prepared i n 9 l i t r e Pyrex solution bottles, each containing 2 l i t r e s of medium. These bot t l e s , placed in a horizontal p o s i t i o n which ensured a large surface area for the culture, were incubated on the bench top f o r 3-g- days. At the .end of t h i s , time, the cultures were f i l t -ered through # 1 Whatman f i l t e r paper, pooled, and s t e r i l i z -ed by Seitz f i l t r a t i o n . The pale yellow f i l t r a t e thus ob-tained was designated type "C" toxin. 14 -However, as mentioned earlier, the f i l t r a t e s prepared according to this original protocol lacked potency, and so some of the possible reasons were investigated. It was thought that perhaps the organism requir-ed a training period in which to adapt i t s e l f to the new medium, and so i t was passed through a series of 7 subcul-tures, with a 24 hour incubation period for each. From . the blood plate, i t was inoculated into a small flask con-taining 60 per cent proteose-peptone broth and 20 per cent C medium, then into one containing 60 per cent proteose-peptone broth and 40 per cent C medium,and so on, with 20 per cent increments, u n t i l f i n a l l y i t was growing on 100 per cent C medium. Two additional subcultures were made, and the f i n a l culture was employed for the inoculat-ion of one of the large bottles* The resulting f i l t r a t e gave a negative cat reactionjin doses up to 3.5 ml. In order to ensure a young viable culture of 12069 alpha whichcontaire d only organisms that had reveal-ed themselves capable of speedy adaptation to growth in C medium^  and which might have retained their toxin-producing a b i l i t y , numerous, rapid subcultures were made. From a 24 hour culture on a blood agar plate, 12069 alpha was inocul-ated into ^ 20 ml. of C medium. 3?our subcultures were made each day for 3 days (that i s , approximately every 3 hours during the day) At the end of this time, a larger glask was inoculated and the regular procedure carried out, but again, no enterotoxin was detected by cat tests. Because other workers (13) had noted the effect of the volume of medium in a container upon the potency of toxin, the next possible factor considered was that of flask size. The use of 1 l i t r e of medium i n a gallon bot-t l e and 800 ml. in a 2 - l i t r e flask made no detectable dif-ference, although these ratios had previously been found favourable for toxin production (16). However, when 300 ML. of C medium in a 2 - l i t r e erlenmeyer flask was employed, a positive cat reaction was obtained from 2.5 ml. of the resulting f i l t r a t e . Thus, i t became evident that the ratio between volume of medium and the surface exposed to the a i r , or in short, the degree of aeration, was a c r i t i c a l point. Therefore, the former method of the aeration of cultures by rotation was resumed. Vacoliter jars of the \ l i t r e and l i t r e size were employed with 100 and 200 ml. of' C medium respectively. They were tied to^the frame of a ba l l m i l l , and rotated on their long axes. The very f i r s t batch prepared in this fashion produced a C toxin with a cat M.R.D. of 1.5 ml., and so the necessity for aeration was established. Further experiments revealed that the pre-sence of uracil exerted no detectable influence on f i l t r a t e s , and so i t was dropped from the formula for C medium. The capacity of the frame, the small amounts of medium which the Vacoliter jars would hold in the horizontal p o s i t i o n ? 16 -and the failure of these jars to withstand autoclaving more than 6 or 8 times - necessitated the alteration of the frame and the procurement of large Pyrex flasks. At the present time, the rotating apparatus handles four 9-litre flasks containing 2 l i t r e s each; that i s , each bat-ch consists originally of 8 l i t r e s of medium - a quantity which seems to u t i l i z e the available f a c i l i t i e s to their capacity. If future expansion of equipment should enable the handling of larger batches of toxin at one time, the problem of aeration might be handled with greater; f a c i l i t y by bubbling air through the cultures. With the adoption of this method, much larger volumes of medium could be dispensed into each flask, and the irksome task of tying the flasks onto a frame would be eliminated. However, un-less the available a i r pressure was controlled with great-er regularity than during the past year, i t s u t i l i z a t i o n would lead to complications. The use of tanks of compress-ed air would provide a more reliable flow, but the expense involved might present a serious handicap. To summarize this aeration problem, one may say that, from time to time, the staphylococcus strain 12069 alpha seemed to undergo some inherent change which render-ed i t incapable of producing enterotoxin under previously favourable conditions. The results obtained suggested that the need for aeration was not completely eliminated by the 17 -use of ur a c i l , and that this simplification in enterotoxin production was inadvisable. Shortly after the resumption of the rotation method of aeration, an apparent change was noted in the gr-owth characteristics of 12069 alpha. At the end of 3^ - days, the culture f l u i d was a distinct yellow; whereas, when grown by the bench top method (with urac i l ) , the cultures became, at most, a pale cream. Moreover, the viscosity of the cul-tures seemed to increase, growth was very heavy, and the resultant f i l t r a t e was of a syrupy consistency. This la t -ter change did not occur immediately, but seemed to devel-op gradually over a period of time with subsequent batches. , It could be correlated neither with any changes in the com-position of the medium, nor with any alterations in prepar-ation. At one time, i t was thought that there might be some relation between the increasing amounts of sunlight and the "sliminess" of the growth; however, subsequent, bat-ches grown during 3^- days of dull light, when bright sun-shine was not evident, were just as turbid as those grown during days which included bright sunshine. About a year ago, the organism began to display another growth variation on G medium. During the f i r s t one, or even two, days of incubation on the rotator - i t assumed a flakey growth which gradually resolved i t s e l f into a heavy, granular, and f i n a l l y a thick rope stage. The large clumps of, growth could be dispersed with vigorous 18 -shaking, hut after a few.^  minutes of continued rotation, they would again "begin to aggregate u n t i l , within an hour or so, the situation would he much as before. Again, no plausible explanation could he offerred. Of course from time to time, new bottles of Bacto-Casamino acids, glucose, etc., had to be opened, hut never did any recorded change in growth seem to coincide with the u t i l i z a t i o n of these "new11 ingredients. During a l l these degrees ofjdhange, there were no colonial variations of 12069 alpha as revealed by growth on blood plates, there was no change in the production of alpha-and enterotoxin on proteose-peptone semi-solid agar, there was no evidence of bacteriophage activity on plates, and there was no marked reduction i n the enterotoxin poten-cy of the C f i l t r a t e s , but there did sesm to be a growing tendency for 2.0cc. rather than 1.5 cc. to constitute the M. R. D. dose. Unfortunately, the increased viscosity of the cul-tures did much to hamper the ease with which C f i l t r a t e s could be obtained. The preliminary f i l t r a t i o n of the cul-ture f l u i d through # 1 Whatman f i l t e r paper no longer prov-ed sufficient treatment prior to the Seitz f i l t r a t i o n . This procedure was, at best, never a rapid one, but with the increasing viscosity, i t became vi r t u a l l y impossible. The pad on a 2- l i t r e Seitz would become completely blocked after less than 1 l i t r e of culture had been s t e r i l i z e d . The 19 use of clarifying pads prior to s t e r i l i z a t i o n , the centri-fugation of the culture f l u i d , the additional use of f i l t e r paper, cheese cloth, etc., and the use of increased pres-sure (15 pounds) which necessitated the inclusion of a sec-ond EK pad, were a l l investigated with varying degrees of success. The clarifying pad treatment, contrary to the early indications, now seems to he of l i t t l e value. At present, the most successful method seems to consist of the follow-ing: f i l t r a t i o n of culture through No. 1 Whatman f i l t e r paper, centrifugation of turbid suspension in 250 ml. cent-rifuge cups for -g- hour at 200 r.p.m., and Seitz f i l t r a t i o n . Even with this treatment, the process i s a slow one, and frequently pads have to be changed a number of times dur-ing the operation. It i s not surprising that this stage of the preparation of enterotoxin constitutes a real bottle neck; It is suggested that possible c l a r i f i c a t i o n of the suspension by some adsorbing agent such as charcoal should be investigated. The readiness with which enterotoxin i s adsorbed by such treatment at later stages of purification does not augur well for this method, but the possibility that some adsorbing agent might be found with a preferent-i a l action on unwanted impurities should not be discredit-ed. III. PURIFICATION Off TOXIN A. Dialysis One of the principal considerations behind the development of C medium was that i t s constituents be dialyz-20 -able, and because this situation was fundamentally ach-ieved, i t i s not surprizing that the f i r s t step in the purification of G toxin should be dialysis. In this way not only the constituents of the medium, but also certain dialyzable products of bacterial growth can be removed. Cold running tap water was found to be a suitable dialyzing agent (12) A 10 gallon granite-ware p a i l was used, the level and inflow of water were controlled by a float value, and the outflow was achieved by 2 suction tubes. By means of this apparatus, approximately 6 l i t r e s of wat-er were allowed to flow in and out of the tank each minute. The C toxin was dispensed i n approximately 500 ml. quantities into the dialyzing tubing (Visking cellulose sausage tubing, "No-Jax", 36/32). This tubing, cut in 3 metre lengths, was knotted at one end, and attached at the other to a rubber stopper carrying a vent tube and a small funnel to f a c i l i t a t e f i l l i n g . This tubing was then clamped by both ends to a rod fixed above the tank - so that the f i l t r a t e was suspended i n the form of a "U". In this way, up to 8 l i t r e s of C toxin could be dialyzed at one time. At the end of 24 hours ( the customary dialyz-ing time), the material was removed from the tank, pooled, and sterilized by Seitz f i l t r a t i o n . The ste r i l e f i l t r a t e of dialyzed C toxin was designated MDC toxin". Casselman (12) i n a study of the efficacy of d i a l -ysis in the removal of medium constituents obtained results - 21 -that indicated the "essentially complete" discharge of C medium. (Table # 1 constitutes a typical i l l u s t r a t i o n of this fact.) Originally, Casselman (12) concluded, from a study of the effect of time of dialysis upon the removal of total solids ( Table # 2), that 24 hours would be the optimum dialyzing time. It was f e l t that the relatively small re-duction in total solids obtained by an additional 24 hours of treatment was not warranted, because of the danger of contamination and the possible destruction of toxin under such conditions. However, Wood (15) found, i n subsequent studies, that neither of these po s s i b i l i t i e s was realized by the increased dialysis period, a.nd that the further re-moval of impurities, as indicated by total S3 l i d s , was larger than original experiments indicated. Thus, at the beginning of this investigation, the 48 hour period was employed; but when growth was observed on one or two occas-ions in the casing, the 24 hour treatment was again adopt-ed. It was considered that by-products of the growth of contaminants would pose a more serious problem than the minute amounts of diffusible medium constituents that would be removed by prolonged dialysis. Because complete data regarding the degree of purification achieved by each processing step was essent-i a l , information concerning the efficiency of dialysis was collected for each batch of toxin prepared, and a few 22 TABLE It The completeness of removal of type C medium by  42*5 hrs. dialysis through Visiting "No-Jax" tubing Before After Dialysis Dialysis ' " . Expt. 1 Expt. 2 Total solids 6.205$ o .oo4$ 0.002$ Amino acids (a ) v. f t . t r . 0 Reducing sugars (b) 0 0 Phosphate (c) t r . f t . t r . Ferrous iron (d) 0 0 a) by triketohydrindene hydrate and by p-dime thy lamino-benzaldehyde b) by Benedict's reagent c) by ammonium mo.lybdate and benzidine d) by dipyridyl TABLE l i t The removal of diffusible materials from type C  toxin by dialysis through Visking "No-Jax" tubing. Time of dialysis Total solids 0 hrs. 1.310 gms. 17 0.028 24 0.011 41 0.008 48 0.008 64 0.010 88 0.009 per 25 ml. sample 23 -of the typical results are summarized i n Table 3. The total solids content was calculated from results obtained by drying 25 ml. samples of toxin at 37°C. or (after the arrival of the vacuum oven) at 37"c.in vacuo. The f i r s t 4 sets of figures (lots 58-61) represent results obtained by Wood, and illust r a t e the difference she observed in the effect of 1 or 2 days dialysis. The differences recorded by this worker were not so great; in fact, the average per-centage removal of total solids by 48 hours dialysis was 99.21$, and by 24 hours - 98.34$ But, in addition to the degree of purification accomplished by dialysis* the percentage recovery of enter-otoxin i s also of prime importance. For the removal of total solids by dialysis would be useless, i f i t were found that the enterotoxin was completely lost in the process. Unfortunately the loss of enterotoxin at this stage of th e purification i s considerable (Table 4), the average re-covery being only 47$. There are a number of possible ex-planations for this marked reduction: part of the enterotoxin may be adsorbed onto the cellulose casing, part of i t may undergo denaturation during the procedure (either at the time of the actual water treatment, or when the material i s exposed to aeration during f i l l i n g and emptying of the tubing, and during Seitz f i l t r a t i o n ) , part of i t may diffuse out through the dialyzing casing, or part of i t may be adsorbed onto the Seitz pad. Variations in the results of - 24 TABLE III; Removal of diffusible materials from type C toxin by dialysis through Visking "No-Jax" tubing. Lot # Time of Total i solids * dialysis C toxin DC toxin reduction 58 20 hrs. 5 . 3 4 5 gms. 0 . 1 9 4 gms- 9 6 . 3 7 $ 59 221 II 5 . 2 6 5 n 0 . 1 6 0 II 9 6 . 9 6 $ 60 44 II 5 . 2 7 0 II 0 . 0 4 5 II 9 9 . 1 5 $ 6 1 48 it 5 . 1 9 5 •i 0 . 0 2 9 II 9 9 . 1 4 $ 1 4 8 n 3 . 3 2 4 II "0 .026 •t 9 9 . 2 2 $ 2 48 n 3 . 1 4 0 a 0 . 0 2 0 II 9 9 . 3 6 $ 3 48 II 2 . 6 6 4 it 0 . 0 2 1 II 9 9 . 2 1 $ 4 48 II 2 . 4 4 0 II 0 . 0 2 1 II 9 9 . 1 4 $ 5 48 n 2 . 5 8 1 ti 0 . 0 2 2 t; II 9 9 . 1 5 $ 15 24 II 3 . 0 3 3 it 0 . 0 4 0 H 9 8 . 6 8 $ 16 24 II 3 . 2 4 7 it 0 . 0 3 9 II 9 8 . 8 0 $ 17 24 ti 2 . 8 0 2 II 0 . 0 3 6 II 9 8 . 7 2 $ 18 24 II 2 . 8 7 0 it 0 . 0 3 4 It 9 8 . 8 2 $ 19 24 II 2 . 8 8 6 ii 0 . 0 3 8 H 98168$ 20 24 II 2 . 5 6 5 II 0 . 0 5 0 II 9 8 . 0 5 $ 30 24 II 2 . 7 7 3 ii 0 . 0 5 6 II 9 7 . 9 8 $ 31 24 II 2 . 9 5 0 n 0 . 0 3 8 II 9 8 . 7 1 $ 32 24 II 3 . 0 4 4 ii 0 . 0 5 8 II 9 8 . 1 0 $ 33 24 II 2 . 3 2 0 n 0 . 0 4 6 II 9 8 . 2 4 $ 34 24 II 2 . 6 4 7 II 0 . 0 4 5 It 9 8 . 3 0 $ 35 24 II 2 . 5 6 5 II 0 . 0 6 2 II 9 7 . 6 8 $ pe»- IOO m l . 25 TABUS IV; Comparison, of C and DC toxins with ^ respect to  enterotoxic potency* Lot Time of Total Volume No Dialysis DC x- Cat Test Doses C Toxin DC Toxin Entero-Vol. ml. Tot. Vol. ml. Tot. toxin recovery 60 44 hrs. 5.0 L. 7.6 L. 1.5 3330 3.5 2170 65$ 61 48 ti 4.9 L. 6.4 II 1.5 3280 3.5 1860 56$ 62 44 ti 4.5 ti 5.2 II 1.5 3000 3.0 1730 58$ 63 46 II 5.3 II 8.5 II 1.5 3530 3.5 2430 69$ 64 46 it 4.3 N 6.8 II 2.0 2150 5.0 1360 63$ 1 48" 1.9 II 2.3 II 1.0 1875 2.5 925 49.3 % 2 48 II r 1.9 II 2.4 it 1.5 1280 3.0 800 62.5 $ 3 48 II 1.7 II 2.4 it 1.5 1148 4.0 600 52.3 $ 4 48 •i 1.9 II 2.3 II 2.0 840 4.5 512 62.0 $ 5 48 II 1.0 II 1.4 II 1.2 852 4.0 350 41.1 $ 6 48 it 2.22 II 2.8 II 1.5 1470 4.0 700 47 .6$ 7 72 II 1.7 II 2.3 it 1.3 1340 5.0 465 34.6 $ 8 72 II 1.7 II 2.2 II 1.5 1150 5.0 440 38.2 $ 9 72 II 1.9 II 2.4 it 1.5 1255 5.0 486 38.7$ 10 24 II 2.6 II 3.3 II 1.5 1735 3.5 942 54.3 $ 11 24 II 2.9 It 3.6 it 1.5 1960 4.0 905 46.1 $ 12 24 II 2.8 II 3.1 II 2.0 1400 5.0 629 44.8 $ 13 24 II 2.9 II 3.5 II 2.0 1470 5.0 700 47.6 $ 14 24 it 2.0 It 2.7 it 1.5 1360 4.0 665 49.0 $ 15 24 n 2.9 II 3.5 ii 1.5 1915 3.5 994 51.9 $ 16 24 ti 2.0 II 2.6 it 1.5 1333 4.0 640 48.1 $ 2t> Lot No. Time of Total Volume Cat Test Doses Dialysis C_ DC C To Vol. ml. xin Tot. DC 1 Vol. ml. 'ox in Tot. Entero-toxin recovery 17 24 hrs. 1.4 L. 1.8 L. 2.3 609 6.0 300 49.3 % 18 24 II 2.0 II 2.5 n 2.0 980 4.5 551 56.2 % 19 24 II 2.2 II 2 .70 II 2.0 1100 5.0 540 49.1 % 24 24 II 6.6 II 8.64 it 2.0 3290 5.0 1730 52.5 % 25 24 II 6.0 II 7.2 II 1.8 3360 5.0 1440 42.9 % 26 24 II 6.6 II 8.18" 1.7 3860 4.5 1820 47.2 % 27 24 II 7.0 II 8.5 II 2.0 3500 4.5 1890 54.0 % 28 24 II 7.1 ti 8.5 ti 2.0 3550 5.5 1550 43.6 % 29 24 II 7.4 ti 8.8 i i 1.5 4940 4.0 2200 44.6 % 30 24 n 7.5 n 8.8 i i 2.0 3750 5.0 1760 47.0 % 31 24 it 7.6 n 9.0 II 2.0 3800 5.0 1800 47.4 % 32 24 II 7.6 II 9.1 II 2.0 3800 5.0 1800 47.4 % 33 24 II 9.8 II 9.2 II 1.5 5080 5 .0 1840 36.3 % 34 24 II 7.2 II 8.9 II 1.5 4800 5.0 1780 37.1 % 35 24 it 7J.2 II 8.9 II 1.5 4800 4.5 1980 41.3 % Note &.) lots 60-64 - data recorded by Casselman (12) b) lots 7, 8, & 9 were dialyzed for 72 hrs. because of the failure of the air pressure supply. - 27 dialysis may be brought about by such undetermined factors as the varying temperature of the water, the changing min-eral constituents of the water, and the quantity of mater-i a l being treated at one time. In spite of these serious disadvantages, dialysis is s t i l l considered by most work-ers to be - in many cases - the least destructive and most convenient method to dispose of medium constituents, and after a l l , the average loss of slightly more than 50 per cent of the original potency i s not excessive when one considers that the total solids content has decreased by some 9 8 . 3 4 per cent. B. Concentration Methods Probably because of the extremely high dilution of the toxic factor, i t has been found essential to adopt some means of concentrating the DC fraction before i t s precipitation. In the past, Gasselman and Wood uti l i z e d a » the Desi-vac treatment available at Connaught Medical Re-search Laboratories for the drying of f i l t r a t e s . But be-cause of the inconvenience of packing, express charges and long delays, several attempts were made to evolve a method which could be employed locally, and which would produce results as good as, i f not better than, those ob-tained by shipping the f i l t r a t e s to Toronto. These methods included vacuum d i s t i l l a t i o n , lyophilization, freezing, and adsorption. - 28 -1 . "Desi - vac" treatment Before the investigations of these various meth-ods of concentration are summarized, the "Desi - vac1? treat-ment, along with typical results .will he outlined. From these findings, i t w i l l become apparent why other means of concentration were attempted. The DC toxin, i n serum bottles containing 400 ml. of f i l t r a t e , was shipped to the Connaught Medical Research Laboratories, School of Hygiene Division, Toronto, where the drying was done on Desi-vac equipment under the direct*, ion of Dr. A* M. Fisher. The dried, dialyzed type C toxin was designated as "type DDC toxin". In most cases, before use, the DDC toxin was re-constituted by the addition of d i s t i l l e d water in a quant-ity that resulted in a forty-fold concentration of the original DC toxin. Invariably, a portion of the DDC remain-ed as an insoluble fraction; this was assigned the term "RP", and the soluble f i l t r a t e , the term "RS". One of the c r i t e r i a for the success of the con-centration method was the reaction of the product to the acid precipitation technique of purification. In general, this involved the following. The RS fraction was adjust-ed to pH 3 . 3 with 5 U HCl. The acidified mixture was plac-ed in the refrigerator for 3 hours, and then the floccular precipitate (designated "p") was f i l t e r e d off on Whatman's 29 # 1 paper, dried at 37°C. in air , or at 23°C. in*vacuum drying oven, and weighed. The f i l t r a t e (designated "FP" was adjusted to pH 7.0 with 5 N NaOH• Dilute a l k a l i (pH 8 to 9) was used to redissolve the acid precipitate P, and this procedure again resulted in the formation of a soluble portion (designated "SP) and an insduble portion (designated "IP") CHART I. A summary of the various terms used in the acid  precipitation technique of purifying DDC toxin. DDC toxin I reconstituted SP IP (solubTe) (insoluble) The results of five of the major experiments in-volving Desi-vac treated DC toxin w i l l now be summarized in various tables, in such a way that the average decreas-ing weight per C.T.D. w i l l be evident, and that the av-erage potency recoveries at the various stages can be 30 -calculated. However, in order that the entire sequence of results can her stated at this point - the following calculations may he made for C and DC toxin. C toxin: Average vol./C.T.D. 1.7 ml. average wt. /ml. .02816 gms. .% average wt. /C- T. D. « .047872 " DC toxin average vol. /C.T.D. 4.4 ml. average wt. /ml. .00056 gms. .*. average wt. /C.T.D. =r .002464 " (The average weights per ml. are calculated from the re-sults list e d in Table III, and the average volume per cat test dose - from Table IV ) Similar calculations have been made for the RS fraction (Table V), aid the SP fraction (Table VI). Prom Table V, iiyis evident that the total sol-ids content per cat test dose of the RS fraction has in-creased considerably; tne average weight of .002312 gms. per cat test dose i s more than twice that obtained by Casselman in early experiments. This increase may be in keeping with the greater destruction of potency by Desi-«» vac treatment over the last 2 years. The higher weight per cat test dose found in those f i l t r a t e s ( 2 and 5) 31 -TABLE V: The r e l a t i o n of the s o l i d content of RS material to cat tes t doses* wt. of RS * 0.5382 gms. * 2.0658 «• exp't 1)1.7605 " » 2) 3.5590 » " 3) 2.7630 '• » 4) 1.5620 «' » 5) 3.5920 " Total CT.D.'s 540 2300 660 850 1400 600 1250 average weight /C.T.p. wt ./C.T.D. .000996 gms. .000902 " .002670 » .004160 « .001975 » .002610 » .002870 " .016183 » = .002312 " re s u l t s recorded by Casselman (13) W n i c h f a i l e d to respond to acid p r e c i p i t a t i o n , seems to substantiate t h i s p o s s i b i l i t y . Table T J reveals, as f a r as may be indicated by t o t a l solids content, the degree of p u r i f i c a t i o n that has been achieved by acid p r e c i p i t a t i o n . The range of 120 - 180 gamma of t o t a l s o l i d s per cat test dose i s not too large, when one considers the d i f f i c u l t i e s i n the determination of t o x i c i t y by cat inoculations* [Further-more, the increase over Casselman's o r i g i n a l figure of TABLE VI: The r e l a t i o n of the s o l i d content of SP mat-e r i a l to cat test doses exp't(l) " #(2) (3) « (4) #(5) wt. of SP * 0 .0243 gms. * 0.1066 0.0365 » 0.0674 « 0.0375 Total C I . J . ' s wt./Q.T.D. 405 1776 203 374 312 3070 .*, average weight /C.T.D. 0.000060 gms. 0.000060 gms. 0.000180 «• 0.000180 « 0.000120 " 0.000600 « .000150 " re s u l t s recorded hy Casselman (13) # no pr e c i p i t a t e formed upon acid treatment 60 gamma, may he p a r t i a l l y explained by his acceptance of a diarrhoeal reaction as a po s i t i v e t e s t . A l l l a t e r c a l c u l a t i o n s have been based on vomiting reactions only. Table VII demonstrates the removal of impurit-ies i n each of the various f r a c t i o n s . The fac t that 47 per cent of the potency of C toxin i s l o s t upon d i a l y s i s explains the difference between 94.49$ (100-5.51$), which i s based on solids per cat test dose, and 98.34$, which 33 TABLE VII: The gradual removal of impurities as revealed by t o t a l s o l i d s content per cat test dose. f r a c t i o n average wt./C.T.I). % of so l i d s i n C C .047872 gms. 100.00 % DC .002464 " 5.51 % RS .002312 " 4.83 % SP .000150 » .31 % TABLE VIII: The per cent of o r i g i n a l DDC so l i d s found  i n the RP f r a c t i o n . wt. of DDC wt. of RP % of DDC exp't (1) 2.203 gms. 0.4425 gms 20.1 % (2) 4.876 » 1.3170 « 27.0 % (3) 3.520 « 0.757 « 21.5 % (4) 1.8640 » 0.302 » 16.2 % (5) 5.3720 » 1.780 « 33.1 $ average removal -23.6 °1o i s based on the removal of solids by d i a l y s i s , with no consideration of toxic content (Table I I I ) . The reduction of t o t a l s o l i d s content from 100 per cent (C toxin) to 0.31 % (SP fraction) indicates a p u r i f i c a t i o n of only s l i g h t l y over 300 times. The actual removal of impurities i s much more than t h i s , but the fact i s masked by the re-34 ductionoof potency at the different stages of the proc-ess. This explains the small difference between the sol-ids content of the DC and RS fractions. Because of the removal of 23.6 per cent (see Table VIII) of the solids of DDC in the form of RP, one would expect a larger var-iation. Table VIII provides a breakdown of the figures in the 5 experiments which supply the information that an average of 23.6 per cent of the solids of DDC are insoluble. But the main purpose for their inclusion here, is so that the relatively higher proportions of insoluble DDC in experiments 2 and 5 can be noted. The fact that these 2 batches were the ones that produced no precipitate indicates that the greater proportion of removed, insoluble substances may be of importance in determining whether or not the RS fraction w i l l react to acid treatment. Tables IX and X supply summaries of the toxin recovered in the different fractions. The figures in IX show how those in X were obtained. The detection of only 46.7 per cent of the original toxin in RS and 3.24 per cent in RP gives a total recovery of only 49.94 per cent of the potency in DC. Thus, the destruction of toxin by the MDesi-vac w treatment has averaged more than 50 per cent. The fact that the average potencies of the SP (15.6 per cent) and the PP (18.4 per cent) fractions give a total of 34 of the original 46.7 per cent of the RS 35 TABLE IX; Determination of average percentage of DC cat test doses recovered in the various fractions. Total cat test doses in the following: DC RS RP SP IP FP exp't (1) 1500 660 60 203 0 300 (2) 2200 (850) (50) (no pp't) 0 (500) (3) 2700 1400 75 374 0 630 (4) 1500 600 50 312 0 120 ( 5 ) (3200) (1250) :(CO> (no-Hp'*) 0 $.050) Totals 5700 2660 185 889 0 1050 % of DC 100 46.7 3.24 15.6 - 18.4 *. The figures for lot s (2) and (5)nwere excluded. TABLE X: The percentage of the original C toxin C.T.D.* detected in the various fractions « fraction C DC RS RP SP PP total C.T.D.'s 12,127* 5,700 2,660 185 889 1050 % of original 100 47 .0 21.9 1.53 7.33 8.66 This figure calculated from average recovery of 47$ toxin after dialysis. 36 f i l t r a t e , shows that the loss due to acid precipitation is not excessive - beingj[just slightly over 10 per cent. Moreover, without the limitations of the cat test, i t i s quite probable that this margin could be further reduced. One of the striking revelations of this table i s the relatively high potencyjof the f i l t r a t e after the acid precipitate has been removed: over 18.4 per cent of the original DC t i t r e is indicated here. Because of this fact a number of experiments to recover this toxin have been attempted and their results w i l l be outlined later. Table X covers the same material as Table IX except that in i t , the percent toxic content of the various fractions i s based on C material. These figures, for the sake of cla r i t y are repeated in an outline chart of the process, (chart II) The recovery of 7.33 per cent of the original toxin, in a degree of purity that seems relatively con/ stant, shows that this means of purifying enterotoxin i s quite feasible. However, the loss of over 50 per cent of the DC toxic t i t r e by Deei-vacv*treatment, i s a serious one, especially when one considers that this process achieves nothing more than concentration of the f i l t r a t e . But a far graver fault l i e s i n the fact thatjthe DDC toxin concentrated by this means does not respond consistently to acid precipitation*. On two occasions (experiments 2 and 5), when large batches of DDC toxin were being treated, the acid precipitate did not form. Moreover, 37 CHART II: The per cent of the original potency of C  toxin to he found in the various fractions  involved in acid precipitation. EP (1.53$) (pp't) C toxin (100$) dialysis DC toxin (47$) MDesi-vac treatment DDC toxin reconstituted RS (21.9$) (f i l t r a t e ) acid ppt'n P (PPH) PP (8.66$ ( f i l t r a t e Sf (7.33$) (fi l t r a t e ) redissolved IP (0$) (PP't) 38 there were no variations in procedure during these five experiments that would account for these negative results. At one time i t was thought that the concentrat-ion of the toxin in the f i l t r a t e might present a c r i t i c a l point. Weak support for this theory may he found in the percentages of the original DC potency found in different RS fractions. ( Table XI ) In experiments (2) and (5) the recovery shows a tendency to be slightly lower than average. However, a stronger indication of this possib-i l i t y was found in the high proportion of toxin in 3?E that would not respond to acid treatment. But, after IT fractions had been concentrated to 3/4, 1/2, and 1/4 of their former volume (by drying at 23'C. in vacuo) they s t i l l failed to reveal addition precipitation. TABLE XI: The percant of the original DC potency recover-o ed in individual fractions. Total cat test doses DC toxin RS fraction % Recovery Casselman(13) 501 3Q1 60 exp't (1) 1500 660 44 (2) 2200 850 39 (3) 2700 1400 52 (4) 1500 600 40 (5) 3200 1250 39 39 Because of the possibility that variations in the'besi-vac" treatment over a period of time might have changed the nature of the DDC material, and because i t was believed that the acid precipitation phenomenon was not necessarily the demonstration of an iso-electric point - small quantities of RS material were treated over a range of pH's. However, of the pH's: 1.5, 3.5, 4.5, 5.5, 7.0, 11.0, and 12.0 - only pH's: 1.5, 3.5, and 12.0 revealed any turbidity at a l l . But the precipitates ob-tained by the pH 1.5 and 12.0 treatments revealed no toxicity (at least i t was 3e ss than 1/4 of the usual concentration) . From time to time, the extreme variation in appearance of the bottles of DDC was noted. Bottles of the same batch, which had supposedly been treated in the same way, were returned with the DDC material in the form of a dark brown gummy mass, voluminous f l u f f y white floccules, a finely-granular yellow precipitate, or some intermediate form of these extremes. Inquiries as to possible variations in Desi-vac treatment, revealed that no known changes had been introduced. The data concern-ing a typical run was as follows: "The temperature of the enterotoxin was not higher than -20fC. u n t i l 64 hours had elapsed. At that time the temperature began to rise and by 70 hours had reachedk temperature of 42aC. It was maintained at that 40 temperature for approximately 7 hours - that i s , u n t i l the completion of the drying. The temperature of the circulating water in the shelves of the vacuum chamber was maintained at 12fiC. for the f i r s t 8-^ -.hours. 10-|-hours after the commencement of the drying, the temperat-ure of the circulating water was raised to ZO^C* and held at that temperature for a period of 13-^  hours;. It was then held at a temperature of 40C'c. for a period of approximately 24 hours, and then at 50*C» u n t i l the drying had bbeen completed.11 (20). Nevertheless, vastly differing products were obtained from theisatme material, and so the problem was investigated. The f i r s t property to be studied was that of weight. The rubber caps were removed from 2 bottles containing the DDC material from 400 ml. of lot#26 DC. The dried materials were transferred immediately to beakers and weighed. The light, white, f l u f f y material weighed 2.3242 gms.; whereas, the deep cream, flaked material weighed only 0.0449 gms. Because i t was thought that this marked difference might be due to the degree of dryness, the materials were placed in the vacuum oven at 50'C. After 96 hours the reduction in weight of the white material was 0.1455gms. and that of the yellow material was 0.0137 gms. Thus i t seemed evident that water content might be influencing the weights to a small extent, but not to a degree that would account for one 41 weight being 5 times that of the other. Additional exper-iments of this type, involving longer drying periods, merely substantiated these earlier results. The next property to be investigated was that of toxicity. The DDC fractions were reconstituted by the addition of d i s t i l l e d water in the usual proportions, and each material displayed a different reaction to the solvent. Most of the white material disappeared almost immediately into the water; whereas, the yellow material required vigorous sti r r i n g before evidence of any dissol-ution was apparent. The white material was, in general found to contain less than 1/5 of the potency of the yellow; and indeed several bottles were found to contain no detectable po/ttency at a l l . Table XII summarizes the results of a comparison of 4 varying bottles of #26 DDC, and from this i t i s apparent that the different response to acid treatment, the.varying pH s t a b i l i t y , and above a l l the irregular retention of potency, indicate that the "Desi-vac" treatment of DC toxin i s most unreliable. 42 TABLE XII; A comparison of four bottles of #26 DDC toxin  which had undergone "Desi-vac" treatment. wt. of 400 ml. original pH * response to pH3.5 pH after 3 hours** original C.T.D.'s C.T.D. in RS 1) 1.672gms * 8.45 PP't 6.3 80 0 2) 0.145 8.65 no pp't 6.0 80 20 3) 0.068 » 8.70 no pp't 6.0 80 20 4) 0.056 » 6.90 PP't 3.2 80 36 * after reconstitition by addition of d i s t i l l e d water, following adjustment to pH 3.5. Descriptions of DDC materials: 1) white, f l u f f y , floccules, very voluminous. 2) light cream, flakey precipitate. 3) deep cream, finely-granular powder. 4) dark brown flakes, with a few almost black specks. -43-2. Vacuum d i s t i l l a t i o n The p o s s i b i l i t y that vacuum d i s t i l l a t i o n might he u t i l i z e d f o r the concentration of enterotoxic f i l t r a t e s was r e a l i z e d early i n t h i s investigation. Such a procedure would he similar to the i n i t i a l step used hy Heidelberger (18, 19) in the p u r i f i c a t i o n of pneumococcal polysaccharides. The experiments were performed on a very small scale, because i t was considered essential to continue shipping material to Toronto f o r Desi-vac treatment (which was known to provide f a i r l y s a t i s f a c t o r y r e s u l t s ) , and be-cause i t was f e l t that small-scale t r i a l runs would at least indicate whether or not the method was applicable to the enterotoxin problem. A 500-ml. Claissen f l a s k , with a c a p i l l a r y pip-ette for the controlled inflow of a i r , was connected through i t s condenser to a drying tube and a Hyy/ac- pump. A 500-ml. d i s t i l l a t i o n f l a s k , with cold water flowing over i t s entire surface, was used as the condenser, and a small, thermo-s t a t i c a l l y controlled water-bath maintained the d i s t i l l i n g f l u i d at a constant temperature. The f i r s t attempted run was performed with the eqipment assembled as above, but with a water pump, instead of the Hyvac, to reducre the a i r pressure. However, i t was soon found that the water supply was neither adequate nor constant enough to s a t i s f y require/* ments. -4-4-In the f i r s t d i s t i l l a t i o n experiment, 400 ml. of DC toxin was treated u n t i l a s l i g h t l y gummy, dark-brown product was obtained* 10 ml. of d i s t i l l e d water was added to t h i s , and the mixture was s t i r r e d frequently, and l e f t at room temperature fo r 2 hours. The insoluble portion (RP) was removed by f i l t r a t i o n , dried at 37*C., and weighed on the f i l t e r paper; the soluble portion (RS) was adjusted to pH 3.3 with IN" HCl. A f l o c c u l a r p r e c i p i t a t e appeared, and the a c i d i f i e d mixture was placed i n the r e f r i g e r a t o r (approximately 5°C.) f o r 3 hours. At the end of t h i s time, the p r e c i p i t a t e (P) was removed by f i l t r a t i o n , dried at 37°C. and weighed on the f i l t e r paper. The f i l t r a t e was adjusted to pH 7.0 with IN NaOH, and stored i n the r e f r i g -erator. Ten ml. of pH 9.1 d i s t i l l e d water was added to pr e c i p i t a t e P on the f i l t e r paper, and passed through the f i l t e r repeatedly f o r some 2 hours. ( L i t t l e or none of the deep brown p r e c i p i t a t e seemed to d i s s o l v e ) . After t h i s treatment only 8 ml. of f i l t r a t e were l e f t . Because one of the c r i t i c a l comparisons of t h i s •I . » ' method of concentration, as against that of Desi-vac treat-ment, would be the amount of material p r e c i p i t a t e d by acid treatment, the various p r e c i p i t a t e s were weighed, and a summary of the calculations follows: wt. of sol i d s i n 400 ml.#2 DC toxin(Table 5)=0.0816 gms. wt. of RP sol i d s =0.0204 * - 4-5 -wt. of RS = 0.0612 gms. wt. of P s ' 0.0041 " wt. of P +• f i l t e r paper = 0.8065 « wt. of ins o l . p + f i l t e r paper = 0.8079 » wt. of soluble P ' ' ? Possible explanations for this phenomenon in-clude the fact that increased humidity produced a "moist" f i l t e r paper which consequently weighed more, or that the hygroscopic nature of the P fraction did not permit the complete release of water molecules after the elution treatment. Calculations for cat doses were complicated by the fact that the amount of p which had dissolved appear-ed to be a negative value. However, for a starting point, cat doses were based on the improbable assumption that a l l the P had dissolved: wt. of solids in 8 ml. of soluble P fraction ; 0.0041 gms. wt. of solids i n 1 ml. = 0.00052 " Wood (12) found 1 M. R. D. V 0.00006 " 1 M. R. D. should be contained in .115 ml. However, when the approximate equivalent of this quantity was inoculated into a cat, a negative reaction was obtained. La.ter inoculations of 10 and 30 times this amount produced no response. Positive cat reactions were obtained from the unprecipitated fraction of RS, and a l -though accurate calculations were impossible, i t seemed apparent that - i f the original total of 133 M. R. D.'s in 400 ml. of # 2 DC toxin - approximately 35 had been le f t behind i n the f i l t r a t e after acid precipitation. There-fore, i t was evident that, somewhere i n thejprocess, a con-siderable fraction of thejtoxic factor had been destroyed -or at least changed in such a way that i t no longer res-ponded to acid treatment as before. A second experiment was carried out to determine at which stage the enterotoxin was lost. However, this time the centrifugation, rather than the f i l t r a t i o n meth-od of separating precipitates was used. The procedure was the same as before except that a DC toxin of slightly higher potency was used ( M. R. D. e 2.5 ml.),and the water bath was maintained at 32*C. instead of 37°C. The centrigugation was carried out in a centrifuge which was cooled by air which had passed through a hose coiled i n a solution of ice and salt. By this means, the temperature of the body of the centrifuge which had an original read-ing of 16.5°C. was kept down to 21°C after one hour of operation, and 28°C. at the end of Z\ hours. In this sec-ond experiment involving vacuum d i s t i l l a t i o n , the precipit-ate formed as before; i t appeared as a dirty yellow, finely granular precipitate which rapidly became floccular. This time the precipitating suspension was placed in the ref r i g -erator for only one hour before further treatment, because i t was thought that the three hour period of acid treatment might be partially responsible for the potency loss. - 4-7-A summary of the results and calculations follows: wt. of solids in 400 ml. of #1 DC = 0.1056 gms wt. of RP - Z 0.0055 » •Ywt. of RS Z 0.1001 " wt. of P S 0.0066 wt of solids in RS after pp't'n " 0.0935 To precipitate (P) was added 10 ml. of pH8.0 d i s t i l l e d water, and the resulting suspension was l e f t at room temperature for one hour. No attempt was made to sep-arate the insoluble fraction by centrifugation, because i t seemed to be negligible. wt. of solids in 10 ml. of P suspension = 0.0066 gms. wt. of solids in 1 ml. of P suspension * 0.00066 M .'• 1 M. R. D. should be contained i n 0.0909 ml. (based on Wood's calculation 1 M. R. D. 0.00006 gms.) Cat inoculations of the equivalent of this amount, 5 times i t , and 10 times i t , gave negative reactions. A study was made of the RS after precipitation: wt. of solids Jinfiltrate after P removed = 0.0935 gms volume at time of test * >|5 ml. •'. wt. of solids in 1 ml. = 0.0187 gms Wood found 1 M. R. D. .005 gms. .*. 1 M. R. D. should be contained i n #0.2674 mis; When the equivalent of this dose was inoculated into a cat, a positive reaction resulted, but half of this dose produced no response; that i s , the presence of at least - 4-S-20 of the original 100 M» R. D.'s was indicated. To the RP fraction was added 10 ml. of d i s t i l l e d water, the suspension was well mixed, and allowed to stand at room temperature for 2 hours,. wt. of RP = 0.0055 gms. 1 cc. contains - 0.00055 But no cat reaction was obtained from 2.5 ml. of the resulting suspension; that i s , the RP fraction contain-ed less than 4 K. R. D.*s CHART HL: Summary of experiment 2 - the effect of vacuum d i s t i l l a t i o n on DC toxin JPC toxiQ-(|.1056 gms.) (100 M. R. D.) I vacuum d i s t i l l a t i o n insoluble fraction (RP) { .0055 gms) ("4 H. R. D.) soluble fraction (RS) (.1001 gms.) acid pp't'n aejyt (p) .0066 gms.) ( 10 M> n» D») \ -•• f i l t r a t e (.0935) (20 M.R.D.) Thus, in this second experiment (Chart ), 20 per cent of the original toxin remained unprecipitated in the RS fraction, no toxin was detected in the acid precipit-ate, and approximately 80 per cent of the original potency of the DC material was unaccounted for. 49 Because of the extreme reduction of toxin in DC material hy vacuum d i s t i l l a t i o n , a third experiment was designed to determine whether the potency reduction in C toxin would he as great. On two occasions, 400 ml. of C toxin and 400 ml. of DC toxin were d i s t i l l e d u n t i l they became dark gummy residues. (The achievement of an absol-utely dry product seemed impossible under the conditions of the procedure; preliminary experiments had revealed that d i s t i l l a t i o n for an additional 24 hours, produced no visi b l e change in the product.) To each of these products was added 20 ml. d i s t i l l e d water, and cat tests were perform-ed on the resultant. These indicated that 66 2/3 per cent of the potency of the C toxin had been retained, but only 25 per cent of the DC toxin. Thus, i t was demonstrated that, under these conditions of vacuum d i s t i l l a t i o n , the impurities found in the C toxin offerred some protection against toxin reduction. The fact that the soluble P fraction of experi-ment 2 gave a positive Biuret, showed that the precip-itate was different in i t s chemical nature as well as in i t s inability to evoke a cat reaction. Because of the aforementioned results, the vacuum d i s t i l l a t i o n method was put aside as being impractical under the present condit-ions. 50 3. Lyophilization. The lyophilization method of concentration of toxin was investigated. The Pacific Fisheries Experimental Station in Vancouver kindly granted opportunities for us to use their lyophilizing plant; and so whenever the ap-paratus was available, material was treated in this way. F a c i l i t i e s were such that 6 l i t r e s of material could he treated at one time. This was dispensed in 750 ml amounts into 2 l i t r e d i s t i l l a t i o n flasks which possessed ground glass stopper joints for attachment to the drying system. The toxin was shell frozen by rotation in a freezing mixture of brine, and then the flasks were stored in the freezing room u n t i l the lyophilization apparatus was free. Unfortunately, f a c i l i t i e s were such that only 1 flask could be shell frozen at a time, and because the process required 20 minutes to 1/2 hour per flask, consider-able time was involved. When a lyophilization run was to be stdrted, the flasks of shell-frozen material were removed from the cold storage room, and attached with a l l possible speed to the outlets which had been previously cleaned and coated with vacuum grease. As soon as the flasks were connected, the pump wasjturned on, and the pressure watched u n t i l i t was evident that there were no leaks. Because there could be no further temperature treatment of the material once i t was attached to the system, i t was most essential that the reduction of pressure be very rapid; in fact, i t was 51 found that i f the pressure did not approach the zero point within five minutes, the material in the flasks would begin to melt and bubble. Then the process had to be stopped, and the material in the flasks refrozen. Usually when this happened, one of the ground glass connections was at fault, but once after lengthy investigation a IJak was found in one of the copper pipes that connected the system to the freezing unit, and once, the behaviour of the pump was the source of trouble. In general, a drying period of 20 to 24 hours was found to be sufficient. This treatment of DC toxin usually yielded a white to cream, crusty precipitate - but occasion-a l l y the product would be light, and floccular. In general i t s reaction to resolution and acid precipitation paralleled f a i r l y closely that of material treated by the "DeBi-vac11 method. However, of 4 batches of DDCmaterial prepared by lyophilization (see Table XIII), the average loss due to the concentration proced ure has been less than 25 per cent instead of the 50 percent reduction by "Desi-vac" treatment. Moreover, so far there has been no variation in the response of the reconstitited material to acid treatment - a potent precipitate of yellow floccules has formed on a l l occasions. 52 TABLE XIII: Determination of the average percentage of Dfe  cat test doses in the various fractions of mat- er i a l treated hy lyophilization. Total cat test doses in the following: •t DC RS RP SP I? PP 1200 900 50 250 25 200 2) 1200 1000 50 230 0 350 3) 1200 870 0 300 0 200 4) 1200 950 0 250 0 250 Totals 4800 3720 100 1030 25 1000 % of DC 100 77.5 2.1 . 21.5 0.5 21 TABLE XIV: The percentage of the original C.T.D.*s in C toxin detected in the various fractions of lyo- philized material. fraction total C.T.D.1s % of original C 10, 212 100 DC 4, 800 47.4 RS -3'.: 720 36.5 RP 100 .9 SP 1, 030 10.1 IP 25 .2 PP .1, 000 9.8 53 Because i t was thought that the lyophilization of C toxin (instead of DC) and the dialysis of the concentrated material might result in a lower.; potency loss, one hatch of f i l t r a t e was treated in this way. Six l i t r e s of C toxin, with an M.R.D. of 2.0 ml., were lyophilized, reconstituted by the addition of 60 ml. of d i s t i l l e d water, placed in Visking i n o cellulose casee, and dialysed for 48 hours. The results are summarized i n TABLE XV. TABLE XV; The percentage of the original C.T.D.'S in the  various fractions of material concentrated by  lyophilization at the C stage. fraction total C.T.D.'a % of original C 3000 100 RS 1260 42 RP 200 6.7 SP 300 10.0 PP 550 18.3 Thus i t i s apparent that this "short cut" in the process does result in a higher retention of toxin in the RS fraction; in fact, there i s almost twice as much : 42 per cent instead of 21.9 per cent. However, the increase i n cat test doses does not correspond to this. Nevertheless, i t i s a higher percentage of recovery (10 per cent instead of 7.33). Unfortunately though, the degree of purity of the acid precipitate seems to be reduced. The calculated weight 54 of solids per cat test dose in the SP fraction i s 960 gamma, which i s considerably higher than the average of 150 gamma revealed by "Desi-vac" treated material. Moreover, this fraction gave a positive Biuret. However, i t i s quite possible that bjs attention to details such as dialysis time, these im-purities could be at least partially removed prior to acid treatment• A study was made of the possibility that the lyo-philization of unsterilized C toxin might present a practical means of eliminating the f i l t r a t i o n , centrifugation, and Seitzing of such large volumes of f i l t r a t e . Six l i t r e s of a 12069 alpha 3$ day culture on C medium were taken down to the Fisheries Research Station, and immediately frozen. Although the imputities seemed to offer some degee of protection for the toxin, because over 50 per cent of the original C.T.D.'s were recovered in the Seitzed f i l t r a t e of the reconstituted material - the relatively large quantity of d i s t i l l e d water that had to be added to a c h i e v e any workable solution at a l l reduced thejoriginal concentration greatly. Moreover, the re-sults after dialysis and acid treatment were much the same as above; namely, a high total solids content per cat test dose, and a positive Biuret reaction from the soluble part of the precipitate. Prom these results, i t may be seen that the lye-philization process seems to offer a better means of concen-tration than does the"Desi-vac" treatment. However, i t must be 55 remembered that the number of batches treated in this way has been small, and that originally the "Desi-vac" treatment also seemed to furnish reliable results* 4* Comparison of Concentration Methods The results obtained from the various methods of concentration have been briefly compared throughout the fore-going discussions, and so for purposes of this section - a summarizing table (XVI) of average results w i l l be offerred. TABLE XVI: A comparison of the results obtained from 3 methods of concentrating DC toxin % recovery of toxin in: SP fraction C_ DC RS RP SP PP wt ./C.T.D. Biuret 1) 100 47 21.9 1.5 7.3 8.7 •wt/I50*. 2) 100 47 ? 2.0 0 0 neg. reaction / 3) 100 47 36.5 0.9 10.1 9.8 240 * Treatment: 1) "Desi-vac" 2) Vacuum d i s t i l l a t i o n 3) Lyophilization 56 C. Precipitation .In addition to acid treatment with HCl, various pre-cipitating agents such as ethyl alcohol, acetone, methanol, and cadmium chloride were used in an attempt*to remove the enterotoxic fraction from RS f i l t r a t e s * A number of preliminary small-scale experiments were employed to determine the optimum conditions for the ethyl alcohol precipitation of enterotoxin from RS fractions, and -based on these results- a large-scale experiment was designed to determine how the degree of purity of ethanpl precipitated material compared with that of SP. The procedure was as follows. 40 ml. of RS and 60 ml. of ethyl alcohol were cooled in the refrigerator. When both registered 5°C.» the RS material was poured slowly into the ethyl alcohol. The mixture was l e f t i n the refrigerator for 6 hours, and then the precipitate was f i l t e r e d off, dried, aad weighed, resuspended, and tested. The results are sum-marized in Table XVII; from which i t i s evident that the ethanol-precipitated material had less than one-fiftieth of the activity of the acid precipitate, and that the weight of alcohol precipitate per C.T.D. was more than 3 times that of jk SP per C.T.D. Thus i t i s apparent that ethanol precip-itation either destroys enterotoxin, or precipitates a greater proportion of impurities with i t Because Sit was thought that some of the 8.66% of 57 TABLE XVII: The recovery and degree of purity of ethanol pre- cipitated toxin compared to SP. Total weight Wt./C.T.D. RS material 3.2284 gms 0.0025 " $ of original tox-in recovered 20$ ethanol pp't 1.762 gms. 0.0083 " 0.53$ SP 0.000150gms. 7 .33$ original toxin might he recovered from the PP fraction, ethan-ol precipitations were done on a number of small quantities. Although a precipitate formed each time, no potency could be detected in i t , and a l l that the procedure accomplished was a destruction of approximately 35per cent of the PP potency. Acetone precipitation of enterotoxin yielded slight-ly better results than did ethanol. The fraction of original toxin recovered was 1.3 per cent,instead of only 0.53 per cent, and the weight per C.T.D. was 0.0052 gms. instead of 0.0083 gms. However, the procedure was s t i l l far inferior to the acid precipitation technique as regards both purity and per-centage recovery of toxin. Cadmium chloride was used as a precipitating agent upon RS material. In general, the technique of Lingood (21) was employed; the 1/3 volume of a 5$ saturated solution of cadmium chloride was used, and a pH of 6.5. A white, flocc-58 ular, voluminous precipitate was formed, and tests revealed that the active component had "been removed from the f i l t r a t e . However, residual cadmium in washed, redissolved, and treated suspensions proved too toxic in animal tests to permit the degree of separation of enterotoxin to he determined. Methanol precipitation produced the best results of a l l the agents employed - with the exception of acid. A pH 4.5 buffer of, 0.5M. KH2PO4 - ISfe^ HPC^  was prepared, and 10 ml. of i t , 20 ml. of methanol, and 20 ml. of RS material were chilled i n the refrigerator. The methanol and buffer were then mixed, and the RS material added slowly. This mixture was then chilled at -2*0. for 3 hours, at the end of which time, a fine white precipitate was evident. This was centrif-uged down jtjb i n a chilled centrifuge cup, resuspended i n c h i l l -ed pH 4.5 phosphate buffer for 15 minutes, and- recentrifuged. This latter step was repeated twice and then the precipitate was dissolved by the addition of phosphate buffer at pH 7.4. This solution was found to contain 22 per cent of the original potency of the RS material, which finding compared quite fav-ourably with the 35 per cent demonstrated in acid precipitate. However, the C.T.D. weighed approximately 6 times the average 150-gamma-weight of SP. Because of the partial success of this experiment, this precipitation procedure was tried on RS fractions which would not respond to acid treatment. In no case did a pred-ipitate form, although a high toxin t i t r e was indicated by 59 cat tests on the fraction. Thus i t seemed that the controll-ing factor of acid precipitation also influenced the response of enterotoxin to methanol. When methanol precipitation was attempted of the FP fraction, a precipitate formed, hut i t lacked potency. D. Adsorption It has been found that "Norit A" charcoal removes 50 per cent of the toxic potency from G toxin, 20 per cent from DC toxin, and 100 per cent from RS fractions. However, in spite of many attempts, thejelution of the potent factor from the charcoal has proven unsuccessful. D i s t i l l e d water and saline at pH's of 3.5, 5.5, 7.5, 9.5, and 11.5, have proven unsuccessful at room temperature, 37'C. and 56'C. Ethyl alcohol and acetone were also ineffective. However, the re-cent report (22) of similar d i f f i c u l t i e s encountered i n the elution of an antibrucella factor found in peptones from "Uor-i t " and the discovery that - i n this case- pyridine was the successful eluting agent, sjiggest that pyridine treatment should be attempted. E. Freezing From time to time, bottles of toxic f i l t r a t e s would freeze during storage in the refrigerator, although none of the other materials around them showed any such tendency. But when one beaker containing RS material displayed a clear ice throughout, except for a very small layer of bright yellow-material i n the bottom, the possible applicability of concen-60 tration of toxin by differe n t i a l freezing was suggested. On three occasions, when small quantities of RS material were frozen, cat tests showed that the reduced volumes of f i l t r a t e s contained toxin i n : 20, 20, and 25 times their former concen-trations. Furthermore, these concentrated RS fractions re-sponded well to acid treatment, and the resultant precipitates were shown to he potent. Unfortunately, the quantities invol-ved were so small that accurate calculations of weights were impossible. However, further experiments might reveal the f r -eezing method to be suitable for the concentration of DC mat-eri a l on a large scale. IV NATURE OF TOXIN A. Potency This subject has been dealt with extensively under the various experiments concerned with the concentration of toxin^ because the degree of purity - as revealed by total solids content - has been one of the pricipal c r i t e r i a for the success or failure of various processes. Therefore, at this point,it should be sufficient to say that enterotoxin appears to be only moderately potent. Weight calculations have shown that i t has been purified to the point where 1 cat test dose contains 0.15 mgms. of solid material. The fact that a number of purification experiments gave results that corresponded so closely, seems to indicate that the toxin may be reaching i t s f i n a l stages of purity. The variation between 60 and 180 gamma is not large when one considers the possible 61 sources of error. The relatively large amount of material, in comparison to other toxins, that constitutes a reacting dose - may he explained by the relative insusceptibility of the cat as compared to humans. Unfoitunately, this point w i l l only be satisfied by the accurate determination of the cat to human reacting-dose-ratio. B> Non- Protein Properties A number of properties of enterotoxin have supplied evidence that the potent material i s not protein-like i n nature* The acid-precipitated material yields positive Mo^ -l i c h and negative Biuret tests with a f a i r degree of regular-i t y , and cats may be inoculated a great many times - especial-ly since the use of f i l t r a t e s prepared o^°synthetic medium -before they reveal any degree of resistance to the toxin. Therefore, an attempt was made to demonstrate fur-ther i t s possible polysaccharide nature by means of enzyme experiments. A preliminary series of experiments using crude proteose peptone f i l t r a t e s of 12069 alpha indicated that a treatment at 37°G. for 4 hours worked satisfactorily for the following enzymes at their optimum pH's. pepsin 1.8 trypsin......... 8.5 caroid 7.0 takadiastase.... 7.0 However, a l t h o u g h the destruction fif alpha toxin by pepsin and trypsin, but not by caroid and takadiastase, 62 TABLE XVIII? Milligrams of nitrogen in the various f i l t r a t e s as revealed by micro-Kjeldahls* mgms ff./C.T.D. I II Average C 2.240 1.8400 2.0400 DC 0.100 0.1200 0.1100 RS 0.055 0.0620 0.0580 RP 0.032 0.0500 0.0410 SP 0.004 0.0030 0.0035 could he demonstrated - even i f exposure times were reduced to 15 minutes, negative eat tests resulted from a l l treated f i l t r a t e s . Greatly diluted concentrations of enzymes and the use of DC f i l t r a t e s gave some evidence of the probable polysaccharide nature of enterotoxin, but i t was realized that further experiments would be valueless u n t i l purified '''^  enzyme preparations had been obtained. Supporting evidence for the non-protein nature of enterotoxin has been recently supplied by mici^Kjeldahl^ results which were made possible by the kind cooperation of members of the Provincial Department of Health Laboratories in Vancouver. The results of the readings on 2 different sets of f i l t r a t e s are summarized in table XVIII. Through them are revealed the marked reduction of nitrogen content at each stage of purification and the f i n a l nitrogen concentration in the SP fraction of less than 2 percent (based on the fact that 1 M.R.D. * 180 gamma). 63 SUMMARY 1. The constituents of a synthetic medium and the cond-itions required by 12069 alpha for good toxin produc-tion have been redetermined. 2. Dialysis treatment has been shown to produce a marked loss of potency, and reduction of solids in C toxin. 3. Vacuum d i s t i l l a t i o n experiments indicated an almost complete loss of potency because of this treatment. 4. Investigation of lyophilization revealed that this means of concentration might produce more reliable and satisfactory results than "Desi-vac" treatment. 5. The freezing method of concentration was shown to be applicable to the enterotoxin problem. 6. The elution of adsorbed toxin from "Norit A" was not achieved. 7. Precipitation of enterotoxin by ethanol, methanol, acet-one, and cadmium chloride was achieved, but only the methanol treatment showed any evidence that would war-rant further investigation. 8. The soluble fraction of acid-precipitated enterotoxin was found to have an average weight of 150 gamma per cat test dose, and a nitrogen content of less than 2 per cent. 6 f BIBLIOGRAPHY 1. Dack, G.M., Cary, W.E., Woolpert, 0., and Wiggers, H.» J. Frev. Med., 4, 167. 1930. 2. Pulton, P . , Br. J. Exp. Path., 24, 65, 1943. 3. Woodward, J.M., and Slane$z, L.W., J. Bact., 42, 819, 1941. 4. Kojima, T.» and Kodama, T., Kitasato Arch. Exper. Med., 16, 197, 1939. 5. Dolman, C.E., Can. J. Pub. Health, Sept 1944, 339. 6. Darrach, M«, p.2. "Studies on thepurification of staph, tox." 7. Hammon, W. MeD., Amer. J. Pub. Health, 31, 1191, 1941. 8. Fildes, P., Richardson, G.M., Knight, B.C.J.G., and Gladstone, G.P., Br. J. Exp. Path., 17, 481, 1936. 9. Knight, B.C.J.G., Bioehem. J., 31, 966, 1937. 10. Porter, J.R., and Pelczar, M.J., J. Bact., 41, 173, 1941. 11. Richardson, G.M., Bioehem. J., 30, 2184, 1945. 12. Favorite, G.O., and Hammon, W.McD., J". Bact., 41, 305, 1941. 13. Casselman, W.G.B., "Studies on the preparation, purif-ication, and properties of staphylococcal enterotoxin," 1947. 14. Surgalla, MLJ., and Hite, K.E., Proc. Soc. Exp. Biol, and Med., 61, 224, 1946. 15. Surgalla, M.J., J. Inf. Dis., 81, 97, 1947. 16. Wood, J.E., "Studies on staphylococcal enterotoxin," 1947. 17. Wolf, P.A., J. Bact., 49, 465, 1945. 18. Heidelberger and Dawson, J. Biol. Chem., 118, 61, 1937. 19. Heidelberger, J. Exp. Med, 64, 559, 1936. 20. Fisher, A.M., Personal communication to Dr. C.E. Dolman. 21. Langood, F.V., Br. J. Exp. Path., 22, 255, 1941. 22. Scherhardt, V.T., Rode, L.J., Foster, J.W., and Oglesby, G., J. Bact., 57, 1, 1949.